Brassinosteroids and analogs as neuroprotectors: Synthesis and structure-activity relationships

Article abstract of DOI:10.1016/j.steroids.2011.10.009

We have demonstrated previously that the brassinosteroid (BR) 24-epibrassinolide exerts neuroprotective effects deriving from its antioxidative properties. In this study, we synthesized 2 natural BRs and 5 synthetic analogs and analyzed their neuroprotect

Full text of DOI:10.1016/j.steroids.2011.10.009

Steroids 77 (2012) 91–99
Contents lists available at SciVerse ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Brassinosteroids and analogs as neuroprotectors: Synthesis and structure–activity
relationships
Jihane Ismaili, Martin Boisvert, Fanny Longpré, Julie Carange, Céline Le Gall, Maria-Grazia Martinoli,
⇑
Benoit Daoust
Département de Chimie-Biologie, Groupe de Recherche en Neuroscience, Université du Québec à Trois-Rivières, 3351 Boulevard des Forges, C.P. 500, Trois-Rivières, Québec,
Canada G9A 5H7
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 August 2011
Received in revised form 18 October 2011
Accepted 20 October 2011
Available online 28 October 2011
We have demonstrated previously that the brassinosteroid (BR) 24-epibrassinolide exerts neuroprotec-
tive effects deriving from its antioxidative properties. In this study, we synthesized 2 natural BRs and
5 synthetic analogs and analyzed their neuroprotective actions in neuronal PC12 cells, against 1-
methyl-4-phenylpyridinium (MPP+), a neurotoxin known to induce oxidative stress and degenerescence
of dopaminergic neurons characteristic of Parkinsonian brains. We also tested the neuroprotective poten-
tial of 2 commercially available BRs. Our results disclosed that 6 of the 9 BRs and analogs tested protected
neuronal PC12 cells against MPP+ toxicity. In addition, our structure–activity study suggests that the ste-
roid B-ring and lateral chain play an important role for their neuroprotective action.
Keywords:
Brassinosteroids
Structure–activity relationship
Neuroprotection
MPP+
Ó 2011 Elsevier Inc. All rights reserved.
Dopaminergic cells
Parkinson’s disease
1. Introduction
BRs are being studied intensively to understand their role in
plant metabolism. Their main physiological effects in plants in-
BRs are highly oxygenated steroids isolated from several vege-
tables, including Vicia faba seeds and pollen [1–3]. To date, about
60 natural BRs have been identified, such as 24-epibrassinolide
(1) and homocastasterone (2a) (Fig. 1), both oxygenated in posi-
tions 2, 3, 6, 22 and 23 [4]. Several BR non-natural analogs, such
as 22S,23S-homocastasterone (2b) and 22S,23S-homobrassinolide
(3) (Fig. 1), have been synthesized [5,6]. Owing to their peculiar
structural features, their extremely low abundance in natural
sources and potent biological activity [7], BRs have been the sub-
ject of numerous synthetic efforts [5,6,8–10]. Their 4 contiguous
chiral centers (C-20, C-22, C-23 and C-24) represent major chal-
lenges in the synthesis of these steroids. The methods developed
by McMorris et al. [5], Mori et al. [6] and Brosa et al. [8] are espe-
cially efficient and versatile, allowing the swift preparation of a
variety of 29-carbon BRs. All these protocols use stigmasterol as
the starting product, and their synthetic scheme can be summa-
rized in 2 main reaction sequences: (i) transformation of the
homoallylic alcohol functionality of stigmasterol into 2,3,6- and
3,6-oxygenated moieties, and (ii) standard osmylation of the
22,23-alkene, resulting in 22,23-dihydroxylated steroidal side-
chains.
clude regulation of hormonal balance, activation of protein and nu-
cleic acid synthesis, enzyme activity, growth promotion, and, most
interestingly, increased resistance to unfavorable environmental
factors, stress and diseases (for review see [7]). BRs have also been
reported to exert anti-oxidative actions [11–17]. Exogenous appli-
cation of natural BRs, such as 24-epibrassinolide (1), to plants en-
hances activities of the enzymatic antioxidants superoxide
dismutase (SOD), catalase (CAT) and peroxidase, reducing lipid
peroxidation [11–15]. Oral administration of homobrassinolide
evokes anti-oxidative outcomes in mammals [16]. Recently, we
determined that 24-epibrassinolide (1) modulates SOD, CAT and
glutathione peroxidase (GPx) in mammalian cells [17].
Several neurodegenerative diseases, e.g. Parkinson’s disease
(PD), are associated with oxidative stress [18]. PD is characterized
by the selective degeneration of nigrostriatal dopaminergic neu-
rons, resulting in dopamine (DA) depletion [19]. Numerous studies
have demonstrated that, in post mortem samples of substantia nigra
pars compacta, DAergic neurons exhibit markers of oxidative stress,
such as lipid peroxidation, DNA oxidative damage, and carbonyl
modifications of soluble proteins [20,21].
L-3,4-Dihydroxyphenyl-
alanine ( -dopa), the amino acid precursor of DA, is nowadays
L
the most effective symptomatic treatment of PD [22]. Clinical re-
ports indicate that consumption of V. faba beans and seedlings,
which contain L-dopa [23,24], has beneficial effects on PD patients
⇑
Corresponding author. Tel.: +1 819 376 5011x3349; fax: +1 819 376 5084.
E-mail address: benoit.daoust@uqtr.ca (B. Daoust).
0039-128X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2011.10.009

92
J. Ismaili et al. / Steroids 77 (2012) 91–99
Fig. 1. Structure of 24-epibrassinolide (1), homocastasterone (2), and 22S,23S-homobrassinolide (3).
[25,26]. However, the concentration of
ficient to explain the magnitude of the responses observed in PD
patients [26]. Thus, it raises the possibility that other compounds
L
-dopa in V. faba is not suf-
Scientific, Ottawa, ON, Canada) were ACS-certified, distilled and
dried prior to use. Reactions requiring anhydrous conditions were
conducted under positive nitrogen atmosphere in oven-dried
glassware, and reaction flasks were fitted with rubber septa for
the introduction of substrates and reagents via standard syringe
techniques. Flash chromatography was performed on Merck silica
gel 60 on Siliaflash P60 (0.040–0.063 mm, Silicycle, Quebec, QC,
Canada) under compressed air pressure. Analytical thin-layer chro-
matography (TLC) was carried out on pre-coated (0.25 mm) Merck
silica gel F54 plates (VWR, Ville Mont-Royal, Qc, Canada) or on sil-
ica gel 60 (0.25 mm, Silicycle, Quebec, QC, Canada) and developed
with an acid solution of ammonium phosphomolybdate. 1H NMR
spectra were recorded on a Varian 200 MHz NMR spectrometer
with CDCl3 (d = 7.26 ppm) as reference. 13C NMR spectra were
traced at 50.3 MHz with CDCl3 (d = 77.1 ppm) as reference. For ace-
tate 9, NMR spectra were charted on a Varian 600-MHz spectrom-
eter. The data reflect the following: chemical shift in ppm,
multiplicity (d = doublet, t = triplet, q = quartet, m = multiplet,
br = broad, dd = doublet of doublets, dt = doublet of triplets,
td = triplet of doublets), number of protons, coupling constants in
Hertz and assignation (if possible). IR spectra were recorded by
Nicolet Impact 420 spectrophotometer. Low resolution mass spec-
tra (LRMS) were recorded on an Agilent Technologies GC system
6890 N and mass detector 5973 with helium as carrier gas. High
resolution mass spectra (HRMS) were measured in the electrospray
(ESI) mode on an HPLC 1200 system with a TOF 6210 detector from
Agilent Technologies. Melting points (mp) were recorded on an
Electrothermal apparatus and were uncorrected. If compounds
were recrystallized prior to determination of their melting points,
the re-crystallization solvent is indicated in brackets.
in V. faba may complement the effect of L-dopa. It is known that
V. faba contains BRs such as 24-epibrassinolide (1), castasterone
and brassinolide [1–3]. Recently, we established that 24-epibras-
sinolide (1) is neuroprotective of nerve growth factor (NGF)-differ-
entiated PC12 (neuronal) cells against MPP+-induced toxicity [17].
MPP+ is the active metabolite of 1-methyl-4-phenyl-1,2,3,6-tetra-
hydropyridine (MPTP), a toxin serving extensively to reproduce
PD in animal models [27]. This neurotoxin is known to act as an
in vitro and in vivo oxidant [28]. MPP+ is selectively taken up by
DAergic neurons via high-affinity DA transporters expressed in
NGF-differentiated PC12 cells [29,30]. The neurotoxin is actively
transported into mitochondria where it interferes with mitochon-
drial respiration through complex I inhibition [31–33]. It evokes
elevated levels of reactive oxygen species (ROS) in MPP+-treated
neuronal cells [34–36] and neuronal cell death by apoptosis
[37,38]. We recently reported that 24-epibrassinolide (1) modu-
lates SOD, CAT and GPx activities and reduces MPP+-induced apop-
tosis and intracellular ROS in neuronal PC12 cells [17].
The aim of the present investigation study was to evaluate the
neuroprotective effects of natural BRs and synthetic analogs and
provide new insights into their structure–activity relationships as
neuroprotective molecules in a well-known in vitro model of PD,
NGF-differentiated PC12 cells [29,39]. We followed the protocol
of Brosa et al. to prepare BRs for this study [8]. This strategy was
particularly well-suited for our work since it allowed us to freely
functionalize steroid A and B rings as well as C20–C29 side-chain.
We were, therefore, able to synthesize 7 BRs with different levels of
oxygenation. We demonstrated that some of these molecules were
neuroprotective against MPP+-evoked toxicity. Structure–activity
analysis revealed the importance of lateral chain and B ring func-
tionalization for neuroprotection. We also noted that BR hydroxyl
group configurations were not crucial for neuroprotection. Overall,
our findings clearly indicate that BRs and analogs are new protec-
tive molecules against MPP+-induced toxicity. Therefore, they
might be regarded as novel candidates to investigate the outcomes
of complementary and/or preventive therapies in neurodegenera-
tive diseases.
2.2. (22E,24S)-3a,5-cyclo-stigmast-22-en-6-one (5)
Compound 5 was prepared in 3 steps from stigmasterol (4)
according to Brosa’s method [8]. Under N2 atmosphere and anhy-
drous conditions, 5.25 g (12.1 mmol) of stigmasterol (4) 95% (Acros
Organics, Fisher Scientific, Ottawa, ON, Canada) were dissolved in
60 mL of toluene. To this solution, 13.0 mL (93.1 mmol, 7.7 eq) of
triethylamine were added with a syringe. The mixture was then
cooled to 0 °C and 2.8 mL (36.1 mmol, 3.0 eq) of MeSO2Cl were in-
cluded drop-wise over 10 min. After stirring at 0 °C for 1.5 h, the
resulting yellow solution was diluted with water and extracted
with toluene (3X). The organic layers were washed with saturated
aq. NaHCO3 (2X) and brine, then dried over MgSO4. Evaporation of
the solvent provided 5.84 g (97%) of the mesylate. This step was re-
peated several times, and the yields obtained were 95–97%. 1H-
and 13C NMR data on the crude compound were consistent with
data in the literature [8].
2. Experimental
2.1. General
All reagents, including 24-epibrassinolide (1) and 22S,23S-
homobrassinolide (3), were purchased from Sigma–Aldrich (Oak-
ville, ON, Canada) unless noted otherwise. All solvents (Fisher

J. Ismaili et al. / Steroids 77 (2012) 91–99
93
4.30 g (8.78 mmol) of mesylate were suspended in 80 mL of water
and 240 mL of acetone. 1.09 g (13.2 mmol, 1.2 eq) of KHCO3 was then
added and the suspension was heated under reflux for 6 h to achieve
i-sterol rearrangement. The solution was allowed to reach room tem-
perature and extracted with EtOAc (3X). The organic layers were
washed with water, saturated aq. NaHCO3 and brine, then dried over
MgSO4. Evaporation of the solvent provided 3.34 g (93%) of yellow oil.
This step was repeated several times and the yields obtained were
87–93%. Note that the 1H- and 13C NMR data on the crude compound
were consistent with data in the literature [8].
(0.06 mmol, 0.04 eq) of a 2.5% OsO4 solution in t-butanol. The mix-
ture was stirred at room temperature for 6 days. 1.20 g of NaHSO3
was then added and the reaction was stirred for another 18 h at
room temperature. t-BuOH was removed under reduced pressure,
and the residue was extracted with EtOAc (6X). Combined organic
layers were washed with water, 0.25 M H2SO4 (3X) and brine, then
dried over MgSO4 and concentrated under vacuum. The crude prod-
uct was purified by flash chromatography on silica gel CH3Cl/EtOH
(0 ? 6%) to provide the 2 desired compounds, 2a and 2b.
3.16 g (7.65 mmol) of crude alcohol were dissolved in 80 mL of
acetone. The solution was stirred and cooled to 0 °C in an ice bath.
Jones reagent (8 N–CrO3: 5.33 g of CrO3 were dissolved in 4.4 mL of
concentrate H2SO4, then water was added to obtain a 20 mL solu-
tion) was included dropwise until the mixture became reddish
(ꢀ3 mL). The reaction was stirred at 0 °C for 5 min. and the solvent
was removed under reduced pressure. The green residue was di-
luted with water and extracted with EtOAc (3X). The organic layers
were combined, washed with saturated aq. NaHCO3 (2X) and brine,
then dried over MgSO4. Evaporation of the solvent provided 2.82 g
(83%) of pure 5. This step was repeated several times, and the
yields obtained were 78–83%. mp lit. 98–99 °C [8] exp. (EtOH)
95–97 °C. 1H NMR (200 MHz, CDCl3) d (ppm) 5.16 (dd, 1H, J = 8
and 15 Hz, H-22), 5.08 (dd, 1H, J = 8 and 15 Hz, H-23), 2.42 (1H,
2.4.1. (22R, 23R)-homocastasterone (2a)
81.9 mg (13.4%). TLC: CHCl3/EtOH (9:1) Rf = 0.21. mp lit. 227–
230 °C [40] exp. 223–225 °C. 1H NMR (200 MHz, CDCl3) d (ppm)
4.05 (m, 1H, H-3), 3.66–3.82 (m, 2H), 3.59 (d, 1H, J = 9 Hz), 2.69
(dd, 1H, J = 3.5 and 12.5 Hz, H-5), 2.3 (dd, 1H, J = 4.3 and
12.9 Hz), 0.9–2.2 (m), 0.76 (s, 3H, H-19), 0.68 (s, 3H, H-18). 13C
NMR (50 MHz, CDCl3) d (ppm) 211.9 (C-6), 74.5 and 72.7 (C-22
and C-23), 68.4 and 68.3 (C-2 and C-3), 56.5, 53.6, 52.5, 50.7,
46.7, 46.3, 42.8, 42.6, 40.2, 39.4, 37.7, 36.9, 30.9, 28.8, 27.6, 26.3,
23.8, 21.2, 19.4, 18.8, 13.6, 13.4, 11.9, 11.9. IR (neat,
3493, 3447, 2965, 2940, 2862, 1699. HRMS calculated for
m
(cmꢁ1))
C
29H51O5 (MH+): 479.3737; found 479.3723. Note that the 1H-
and 13C NMR data on this purified compound were consistent with
data in the literature [5].
dd, J = 8 and 14 Hz, H-7a), 2.10–1.10 (m, 25H), 1.03 (s, 3H, H-19),
1.02 (d, 3H, J = 6 Hz, H-21), 0.98–0.78 (m, 9H), 0.73 (s, 3H, H-18).
13C NMR (50 MHz, CDCl3) d (ppm) 210.1 (C-6), 138.3 (C-22),
129.7 (C-23), 57.7, 56.1, 51.5, 47.0, 46.5, 46.3, 45.0, 42.8, 40.7,
39.8, 35.6, 35.0, 33.7, 32.1, 29.1, 26.1, 25.6, 24.4, 23.1, 21.4, 21.3,
2.4.2. (22S,23S)-homocastasterone (2b)
71.2 mg (11.5%). TLC: CHCl3/EtOH (9:1) Rf = 0.29. mp lit. 204–
207 °C [5] exp. 206–208 °C. 1H NMR (200 MHz, CDCl3) d (ppm)
4.05 (m, 1H, H-3), 3.72–3.82 (m, 1H), 3.58–3.62 (m, 2H, H-22 and
H-23), 2.68 (dd, 1H, J = 3.5 and 12.5 Hz, H-5), 2.3 (dd, 1H, J = 4.3
19.9, 19.2, 12.5, 12.4, 11.9. IR (KBr,
m
(cmꢁ1)) 2943, 2856, 1685,
1478, 1369, 1298, 1167, 966, 922. LRMS (m/z, relative intensity)
410 (M+) (100), 392 (10), 367 (70), 298 (70), 271 (75). Note that
1H- and 13C NMR data on this compound were consistent with data
in the literature [8].
and 12.9 Hz), 0.9–2.2 (m), 0.76 (s, 3H, H-19), 0.70 (s, 3H, H-18). 13
C
NMR (50 MHz, CDCl3) d (ppm) 211.9 (C-6), 72.2 and 70.6 (C-22
and C-23), 68.4 and 68.3 (C-2 and C-3), 56.3, 53.7, 52.6, 50.7, 49.6,
46.7, 43.5, 42.5, 42.3, 40.2, 39.4, 37.6, 27.8, 26.9, 26.3, 24.2, 21.7,
21.2, 18.6, 17.7, 14.5, 14.1, 13.5, 11.9. IR (neat,
m
(cmꢁ1)) 3361,
2.3. (22E,24S)-stigmasta-2,22-dien-6-one (6)
2948, 2869, 1713. HRMS calculated for C29H51O5 (MH+): 479.3737;
found 479.3723. Note that the 1H- and 13C NMR data on this purified
compound were consistent with data in the literature [5].
Under inert atmosphere, a mixture of pentacyclic ketone 5
(4.88 g, 11.87 mmol), pyridinium chloride (0.51 g, 5.94 mmol,
0.50 eq), dry LiBr (0.69 g, 5.94 mmol, 0.50 eq) and dimethylacet-
amide (DMA, 50 mL) was heated under reflux conditions for 4 h.
The reaction was then cooled to 0 °C to initiate crystallization.
The pure product was collected by filtration. The filtrate was also
cooled to 0 °C to induce further crystallization, and the product
was again filtered. This step was repeated 2 more times until no
crystallization were apparent. 4.67 g (96%) of pure 6 was obtained.
mp lit. 110–111 °C [8] exp. (EtOH) 111–112 °C. 1H NMR (200 MHz,
CDCl3) d (ppm) 5.66–5.58 (m, 2H, H-2 and H-3), 5.13 (dd, 1H, J = 8
and 15 Hz, H-22), 5.07 (dd, 1H, J = 8 and 15 Hz, H-23), 2.40–1.20
(m, 26 H), 1.04 (d, 1H, J = 7 Hz, H-21), 1.02–0.80 (m, 9H), 0.71 (s,
3H, H-19), 0.69 (s, 3H, H-18). 13C NMR (50 MHz, CDCl3) d (ppm)
212.4 (C-6), 138.3 (C-22), 129.8 (C-23), 125.2 and 124.8 (C-2 and
C-3), 57.1, 56.1, 54.1, 53.7, 51.5, 47.3, 42.9, 40.7, 40.3, 39.6, 37.9,
32.1, 29.0, 25.6, 24.2, 22.0, 21.4, 21.3, 19.2, 13.7, 12.5, 12.4. IR
2.5. (22E,24S)-2a,3a-dihydroxy-5a-stigmast-22-en-6-one (7)
0.25 g (0.61 mmol) of diene 6 was suspended in 7.5 mL of ace-
tone/water (9:1) containing 0.18 g (1.52 mmol, 2.50 eq) of 4-meth-
ylmorpholine N-oxide (NMO). After the addition of 0.20 mL of a
2.5% OsO4 solution in t-butanol (0.02 mmol, 0.03 eq), the suspen-
sion was stirred at room temperature for 12 h. Solid sodium bisul-
fite was then added in excess to destroy the remaining OsO4 and
stirred at room temperature for 5 h. The resulting mixture was di-
luted with water and extracted with EtOAc (3X). The organic layers
were washed successively with 5% NaHSO3 (2X), saturated aq.
NaHCO3 (2X) and brine (2X), then dried over MgSO4. Evaporation
of the solvent gave 217 mg of a yellow solid which was first puri-
fied by flash chromatography on silica gel (CHCl3/MeOH 2%) and
then recrystallized from EtOH to obtain 160 mg of pure 7 (60%)
as white filaments. mp lit. 238–240 °C [41] exp. (EtOH) 237–
239 °C. 1H NMR (200 MHz, CDCl3) d (ppm) 5.13 (dd, 1H, J = 8 and
15 Hz, H-22), 5.07 (dd, 1H, J = 8 and 15 Hz, H-23), 4.04 (m, 1H, H-
3b), 3.75 (m, 1H, H-2b), 2.65 (dd, 1H, J = 3 and 13 Hz, H-5), 2.31
(dd, 1H, J = 4 and 13 Hz), 2.20–1.10 (m, 24H), 1.03 (d, 3H,
J = 6 Hz, H-21), 0.74–0.86 (m, 9H), 0.75 (s, 3H, H-19), 0.68 (s, 3H,
H-18). 13C NMR (50 MHz, CDCl3) d (ppm) 212.3 (C-6), 138.2 (C-
22), 129.9 (C-23), 68.6 and 68.5 (C-2 and C-3), 57.0, 56.1, 54.0,
51.5, 50.9, 47.0, 43.1, 42.8, 40.6, 40.5, 39.6, 37.9, 32.1, 28.9, 26.5,
(KBr,
m
(cmꢁ1)) 2961, 2863, 1710, 1460, 1389, 965. LRMS (m/z, rel-
ative intensity) 410 (M+) (100), 395 (80), 367 (25), 297 (25), 269
(30). Note that the 1H- and 13C NMR data on this purified com-
pound were consistent with data in the literature [5,8].
2.4. General procedure for preparation of compounds 2a and 2b
0.62 g (1.52 mmol) of diene 6 was dissolved in 30 mL of t-BuOH/
H2O (1:1). To this solution was added 1.26 g (9.10 mmol, 6.00 eq) of
K2CO3, 3.00 g (9.10 mmol, 6.00 eq) of K3Fe(CN)6, 0.14 g (0.30 mmol,
0.20 eq) of dihydroquinidine-4-chlorobenzoate (DHQD), 0.29 g
(3.00 mmol, 2.00 eq) of methanesulfonamide and 0.60 mL
25.6, 24.2 21.4, 21.3, 19.2, 13.8, 12.5. IR (KBr,
m
(cmꢁ1)) 3384,
2954, 2856, 1712, 1451, 1374, 1048, 966. LRMS (m/z relative inten-
sity) (TMS derivative) 573 (M+-CH3) (30), 498 (15), 471 (100), 382

94
J. Ismaili et al. / Steroids 77 (2012) 91–99
(30). HRMS calculated for C29H49O3 (MH+): 445.3682; found
445.3676. Note that the 1H- and 13C NMR data on this purified
compound were consistent with data in the literature [41].
resulting clear solution was concentrated under vacuum. The yel-
lowish crude product was diluted with water and extracted with
EtOAc (3X). The organic layers were combined and washed with
water (3X), saturated aqueous NH4Cl (2X) and brine (2X). The or-
ganic layer was dried over MgSO4 and concentrated under vacuum
to afford 447 mg of a yellowish solid. This crude product was fil-
tered over a silica gel pad to eliminate colored impurities with
CHCl3/EtOH (96:4) as eluent. 425 mg of 10 (89%) were isolated as
white crystals. mp lit. 146–147 °C [42] exp. (EtOH) 153–155 °C.
1H NMR (200 MHz, CDCl3) d (ppm) 5.14 (dd, 1H, J = 8 and 15 Hz,
2.6. Homobrassinolide (8)
A solution of homocastasterone 2a (0.106 g, 0.22 mmol) in
CHCl3 (10 mL) was added drop-wise to a stirred solution of peroxy-
trifluoroacetic acid (0.22 mmol, 10 eq) [prepared from 30% aq.
H2O2 (0.26 mL, 0.22 mmol) and (CF3CO)2O (1.58 mL, 1.85 mmol,
5 eq) in CHCl3 (5 mL) at 0 °C]. The reaction mixture was stirred at
room temperature for 2 h, diluted with CHCl3 (10 mL), and the
resulting solution was washed with H2O, saturated aq. Na2CO3
(2X), saturated NaHSO3 (2X) and brine, then dried over MgSO4.
Evaporation of the solvent gave a colorless solid which was recrys-
tallized from EtOAc to obtain 22R,23R-homobrassinolide 8 (70 mg,
65%) mp lit. 249–251 °C [40] exp. (EtOH) 245–247 °C. 1H NMR
(200 MHz, CDCl3) d (ppm) 4.05 (m, 3H), 3.71 (m, 2H), 3.57 (d, 1H,
J = 8.6 Hz), 3.12 (dd, 1H, J = 5.1 and 12.1 Hz), 0.87–2.2 (m), 0.71
(s, 3H, H-18). 13C NMR (50 MHz, CDCl3) d (ppm) 176.2 (C-6), 74.4
and 72.8 (C-22 and C-23), 70.4 and 68.1 (C2 and C-3), 68.1 (C-7),
58.2, 52.5, 51.3, 46.3, 42.5, 41.5, 40.9, 39.6, 39.2, 38.3, 36.9, 31.0,
28.8, 27.6, 24.7, 22.2, 21.2, 19.4, 18.8, 15.5, 13.5, 11.9, 11.7. IR (neat,
H-22), 5.04 (dd, 1H, J = 8 and 15 Hz, H-23), 3.58 (m, 1H, H-3a),
2.30 (dd, J = 5 and 13 Hz, 1H, H-5), 2.23 (dd, J = 3 and 13 Hz, 1H,
H-7), 2.10–1.10 (m, 28H), 1.03 (d, J = 6 Hz, 3H, H-21), 0.88–0.74
(m, 9H), 0.66 (s, 3H, H-18). 13C NMR (50 MHz, CDCl3) d (ppm)
211.2 (C-6), 138.2 (C-22), 129.8 (C-23), 70.9 (C-3), 57.1, 57.0,
56.1, 54.2, 51.5, 47.0, 43.1, 41.2, 40.6, 39.6, 38.1, 36.9, 32.1, 30.9,
30.3, 29.0, 25.6, 24.3, 21.7, 21.4, 21.3, 19.2, 13.4, 12.5, 12.4. IR
(KBr,
m
(cmꢁ1)) 3411, 2946, 2867, 1708, 1459, 1383, 1062, 968.
LRMS (m/z, relative intensity) 428 (M+) (80), 410 (25), 385 (30),
367 (60), 316 (85), 287 (100). HRMS calculated for C29H49O2
(MH+): 429.3737; found 429.3727.
2.9. (22E,24S)-3a-hydroxy-5a-stigmastan-22-en-6-one (11)
m
C
(cmꢁ1)) 3436, 2965, 2933, 2837, 1695. HRMS calculated for
29H51O6 (MH+): 495.3686; found 495.3674. Note that the 1H-
Under inert atmosphere, in a 50-mL round-bottomed flask,
400 mg (0.93 mmol) of alcohol 10 was mixed with 978 mg
(3.73 mmol, 4.00 eq) of Ph3P and 0.623 g (3.73 mmol, 4.00 eq) of
p-nitrobenzoic acid. To this solid mixture were added 20 mL of
anhydrous THF with a syringe. The solution was cooled to 0 °C,
and 0.75 mL (3.73 mmol, 4.00 eq) of diethyl azodicarboxylate
(DEAD) was added drop-wise over 15 min. The orange reaction
mixture was stirred at 0 °C for 20 min and at room temperature
for 24 h. The solution was then diluted with water and extracted
with CHCl3 (3X). The organic layers were combined and washed
with water (3X), saturated aqueous NaHCO3 (2X) and brine (2X).
The organic layer was dried over MgSO4, filtered and concentrated
under vacuum. The crude product was purified by flash chroma-
tography on silica gel with CHCl3 as eluent. 328 mg of nitrobenzo-
ate (61% crude) were obtained as a yellowish solid. 1H NMR
(200 MHz, CDCl3) d (ppm) 8.30 (d, 2H, J = 9 Hz), 8.21 (d, 2H,
J = 9 Hz), 5.42 (m, 1H, H-3b), 2.64 (dd, 1H, J = 4 and 12 Hz, H-5),
2.32 (dd, 1H, J = 3 and 13 Hz), 2.10–1.10 (m, 28H), 1.02 (d, 3H,
J = 6 Hz, H-21), 0.83–0.75 (m, 9H), 0,70 (s, 3H, H-18). 13C NMR
(50 MHz, CDCl3) d (ppm) 211.6 (C-6), 164.0, 150.7, 138.2 (C-22),
136.4, 130.8(2), 129.9 (C-23), 123.8(2), 71.2 (C-3), 57.1, 56.2,
54.3, 53.1, 47.0, 43.1, 41.6, 40.7, 39.6, 38.2, 32.9, 32.1, 29.0, 25.6,
25.3, 24.2, 21.4, 21.3, 19.2, 12.7, 12.5.
and 13C NMR data on this purified compound were consistent with
data in the literature [5].
2.7. (22E,24S)-3b-acetoxy-5a-stigmastan-22-en-6-one (9)
1.80 g (4.38 mmol, 1.00 eq) of pentacyclic ketone 5 were added
to 90 mL of glacial acetic acid. After complete dissolution, 8.8 mL of
aqueous 1 M H2SO4 were added. The heterogeneous solution was
refluxed for 2 h. The solution was then cooled to room tempera-
ture, diluted with water and extracted with EtOAc (4X). The organ-
ic layers were combined, washed with water (6X), saturated
NaHCO3 (2X) and brine (2X). After evaporation of the solvent, the
yellow crude product was dissolved in 20 mL of pyridine.
1.85 mL of acetic anhydride and 18 mg of 4-dimethylaminopyri-
dine (DMAP) were then added. After stirring at room temperature
for 3 h, the orange mixture was diluted with water and extracted
with EtOAc (4X). The organic layers were combined and washed
with water (3X), saturated NaHCO3 (2X) and brine (2X). The solu-
tion was then dried over MgSO4 and concentrated under vacuum
to obtain 1.56 g of 9 (74%) as white crystals. mp lit. 146–147 °C
[42] exp. 142–144 °C. 1H NMR (600 MHz, CDCl3) d (ppm) 5.13
(dd, 1H, J = 8 and 15 Hz, H-22), 5.03 (dd, 1H, J = 8 and 15 Hz, H-
23), 4.66 (m, 1H, H-3
a
), 2.31 (dd, J = 5 and 13 Hz, 1H, H-5), 2.25
320 mg (0.55 mmol) of crude nitrobenzoate was suspended in
30 mL of MeOH. 0.375 g (1.25% w/v) of NaOH was added to this
suspension. The mixture was stirred at room temperature for
14 h. The yellow solution was concentrated under vacuum, diluted
with water and extracted with EtOAc (3X). The combined organic
layers were washed with water (3X), saturated aqueous NH4Cl
(2X) and brine (2X). The organic layer was then dried over MgSO4,
filtered and concentrated under vacuum. The yellow solid obtained
(227 mg) was purified by flash chromatography with CHCl3/MeOH
1% as eluent. 196 mg of 11 (49% over 2 steps) were isolated as
white powder. mp exp. (EtOH) 194–196 °C. 1H NMR (200 MHz,
CDCl3) d (ppm) 5.14 (dd, 1H, J = 8 and 15 Hz, H-22), 5.04 (dd, 1H,
J = 8 and 15 Hz, H-23), 4.17 (m, 1H, H-3b), 2.68 (m, 1H, H-5), 2.25
(dd, J = 3 and 13 Hz, 1H), 2.08–1.08 (m, 28 H), 1.03 (d, J = 6 Hz,
3H, H-21), 0.98–0.72 (m, 9H), 0.68 (s, 3H, H-18). 13C NMR
(50 MHz, CDCl3) d (ppm) 213.0 (C-6), 138.3 (C-22), 129.8 (C-23),
65.7 (C-3), 57.2, 56.1, 54.1, 51.9, 51.5, 47.1, 43.1, 41.8, 40.6, 39.6,
38.2, 32.1, 31.9, 29.0, 28.4, 27.9, 25.6, 24.2, 21.4, 21.3, 19.2, 12.5,
(dd, J = 3 and 13 Hz, 1H, H-7), 2.02 (s, 3H, H-31), 2.03–1.10 (m,
27H), 1.01 (d, J = 6 Hz, 3H, H-21), 0.80 (t, J = 7 Hz, 3H, H-29), 0.78
(d, J = 7 Hz, 3H), 0.77 (s, 3H, H-19), 0.68 (s, 3H, H-18). 13C NMR
(150 MHz, CDCl3) d (ppm) 211.2 (C-6), 170.6 (C-30), 137.9 (C-22),
129.6 (C-23), 72.8 (C-3), 56.8, 56.5, 55.9, 51.2, 46.6, 42.8, 40.9,
40.4, 39.3, 37.9, 36.4, 31.8, 28.7, 26.8, 26.1, 25.4, 24.0, 21.4, 21.3,
21.1(2), 19.0, 13.0, 12.2(2). 1H-NOESY (a correlation was observed
between H-5a (2.31 ppm) and H-3a (4.66 ppm)). IR (KBr, m
(cmꢁ1))
2941, 2876, 1741, 1708, 1478, 1361, 1244, 1023, 960. LRMS (m/z,
relative intensity) 470 (M+) (70), 410 (60), 367 (80), 329 (100),
271 (95). HRMS calculated for C31H50O3Na (M + Na)+: 493.3653;
found 493.3652.
2.8. (22E,24S)-3b-hydroxy-5a-stigmastan-22-en-6-one (10)
525 mg (1.12 mmol) of acetate 9 were suspended in 45 mL of
MeOH. To this suspension were added 2.25 g (5% w/v) of KOH,
and the mixture was stirred at room temperature for 2 h. The
12.4. IR (KBr,
m
(cmꢁ1)) 3290, 2941, 2869, 1707, 1465, 1000, 966.

J. Ismaili et al. / Steroids 77 (2012) 91–99
95
LRMS (m/z, relative intensity) 428 (M+) (20), 410 (80), 367 (60), 316
(55), 271 (100). HRMS calculated for C29H49O2 (MH+): 429.3733;
found 429.3727.
Total cellular LDH was determined by lysing the cells with 1% Triton
X-100 (high control); the assay medium served as a low control and
was subtracted from all absorbance measurements:
ðExperimental value ꢁ Low controlÞ ꢂ 100
2.10. (24S)-3b-hydroxy-5a-stigmastan-6-one (12)
Cytotoxicityð%Þ ¼
:
ðHigh control ꢁ Low controlÞ
100 mg (0.23 mmol) of sterol 10 were dissolved in 2.5 mL of
EtOAc. 20 mg of Pd/C 10% were added to this solution. The reaction
flask was equipped with a rubber septum, and hydrogen gas was
bubbled directly into the solution for 5 min. The solution was stir-
red for 3 h under hydrogen atmosphere at room temperature. The
preceding step (5 min of bubbling, followed by 3 h of stirring) was
repeated twice and stirred for another 14 h under hydrogen atmo-
sphere. The reaction mixture was filtered through Celite. The fil-
trate was washed twice with water and twice with brine. The
organic layer was dried over MgSO4, and concentrated under vac-
uum. 95 mg of an amorphous solid were obtained. This solid was
purified by flash chromatography with CHCl3/MeOH gradient
(0 ? 6%) as eluent to provide 65 mg of 12 (65%) as a white solid.
mp 118–120 °C. 1H NMR (200 MHz, CDCl3) d (ppm) 3.57 (m, 1H,
2.13. Statistical analysis
Significant differences between groups were determined by 1-
way ANOVA, followed by Tukey’s post hoc analysis with the Graph-
Pad Instat program, version 3.06, for WindowsÓ (San Diego, CA,
www.graphpad.com). All data, at the 95% confidence interval, are
expressed as means standard error of the mean (SEM) from 3
independent experiments. Asterisks indicate statistical differences
between the treatment and respective control condition (⁄p < 0.05,
⁄⁄p < 0.01 and ⁄⁄⁄p < 0001); full circles show statistical differences
between the treatment and MPP+ condition
(
ꢃꢃp < 0.01 and
ꢃꢃꢃp < 0001).
H-3a), 2.38 (dd, 1H, J = 5 and 13 Hz, H-5), 2.20–0.82 (m, 45H),
0.68 (s, 3H, H-18). 13C NMR (50 MHz, CDCl3) d (ppm) 211.2 (C-6),
70.9 (C-3), 57.0(2), 56.2, 54.2, 47.0, 46.0, 43.2, 41.2, 39.7, 38.1,
36.9, 36.3, 34.1, 30.9, 30.3, 29.4, 28.3, 26.3, 24.2, 23.3, 21.7, 20.0,
3. Results
19.2, 18.9, 13.4, 12.2(2). IR (KBr,
m
(cmꢁ1)) 2395, 2354, 1716,
3.1. Chemistry
1531, 1416, 1052, 927. HRMS calculated for C29H51O2 (MH+):
431.3889; found 431.3887.
Seven brassinosteroids and analogs were prepared according to
the synthetic pathway depicted in Scheme 1. The common starting
material for all these compounds was ketone 5 (see Scheme 1), eas-
ily prepared from stigmasterol (4) with the modified protocol of
Brosa et al. [8]. Ketone 5 was then reacted with lithium bromide
and pyridinium chloride in refluxing DMA to produce dienone 6
[47]. This reaction consists of acid-catalyzed bromide opening of
the 3-membered ring, followed by HBr elimination. We altered
the protocol of Brosa et al. slightly to improve the efficiency of their
method. Indeed, when preparing dienone 6 from stigmasterol (4)
(with an overall yield of 56% over 4 steps), Brosa et al. performed
2 laborious time-consuming chromatographic purifications. We
eliminated both chromatographic purifications and obtain pure
dienone 6 with only simple crystallization in the last step. The
overall yield for these 4 steps was also improved to 72%. To form
natural 22R, 23R-homocastaterone (2a), we treated compound 6
with Sharpless’ asymmetric dihydroxylation mixture (OsO4,
K3Fe(CN)6, K2CO3, MeSO2NH2, DHQD) [5,48]. A 1:1 mixture of nat-
ural 2a and non-natural 2b was obtained. Separation by flash chro-
matography yielded pure 22R, 23R-homocastasterone (2a) (13%
overall yield) and 22S,23S-homocastasterone (2b) (11% overall
yield). 22R, 23R-homobrassinolide (8) was synthesized by Bae-
yer–Villiger oxidation of 2a with peroxytrifluoroacetic acid to pro-
vide a mixture of 7-oxalactone and 6-oxalactone in a 9:1 ratio from
which pure 22R, 23R-homobrassinolide (8) was isolated by re-
crystallization from EtOAc with 65% yield.
2.11. Cell culture and treatments
PC12 cells were obtained from the American Type Culture Col-
lection (Rockville, MD) and maintained in a controlled environ-
ment at 37 °C and 5% CO2 atmosphere. They were grown in
RPMI-1640 medium supplemented with 5% fetal bovine serum
(FBS), 10% horse serum and gentamicin (50 lg/mL). The culture
medium was changed every 2 days, and the cells were seeded at
a cellular density of 25,000 cells/cm2. Neuronal differentiation
was induced for 4 days with 50 ng/mL of NGF in RPMI-1640 med-
ium supplemented with 1% FBS. To examine the effect of commer-
cially available BRs (1 and 3), synthesized natural BRs (2a and 8)
and analogs (2b, 7, 10, 11 and 12) on MPP+-induced cellular death,
neuronal PC12 cells were pre-treated with BRs (10ꢁ9 M), analogs
(10ꢁ9 M) or vehicle (culture medium) for 3 h and exposed to
MPP+ 5 mM for 24 h while maintaining BR or analog concentration
at 10ꢁ9 M. After dose–response experiments, the final concentra-
tion of 10ꢁ9 M 24-epibrassinolide (1) was chosen in our previous
study as the lowest dose capable of rescuing cells from MPP+-in-
duced cellular death [17]. For comparative purposes, we worked
with the same concentration for all BRs and analogs used in this
study. All experiments were performed in phenol red-free medium
and charcoal-stripped serum to remove steroids.
2.12. Cytotoxicity measurements
We also prepared diol 7 devoid of any OH at positions 22 and
23. Comparing neuroprotective activity of 7 with 2a and 2b al-
lowed us to evaluate the importance of oxygenation at positions
22 and 23. Dihydroxylation of dienone 6 with OsO4/NMO at room
temperature for 12 h generated a 10:1 mixture of 2a,3a-diol 7/2b,
3b-diol. Purification by flash chromatography gave pure compound
Cytotoxicity was evaluated by colorimetric assay based on the
measurement of lactate dehydrogenase (LDH) activity released from
damaged cells into the supernatant [43]. LDH is a stable cytoplasmic
enzyme present in all cells. It is rapidly released into cell culture
supernatant upon damage of the plasma membrane. The amount
of enzyme activity detected in culture supernatant correlates with
the portion of lysed cells [44,45]. NGF-differentiated PC12 cells were
grown and treated in collagen-coated 96-wellplates. Then, 100
LDH substrate mixture was added to 50 L of cell-free supernatant,
as described elsewhere [46]. The plate was incubated, protected
from light, for 20 min. Absorbance was measured at wavelength
490 nm on a microplate reader (Thermolab System, Franklin, MA).
7 with 60% yield.
We then moved onto synthesize analogs of the 3-deoxy (homo-
teasterone) family. Aqueous acetic acid opening of ketone 5 first
gave a 2.5:1 mixture of 9 and 10. Treatment of this crude mixture
with DMAP, pyridine and Ac2O produced 9 with 75% yield. Hydroly-
sis of 9 under basic conditions yielded alcohol 10 with 89% yield.
Note that direct opening of ketone 5 in aqueous basic or acidic con-
ditions (without the presence of any other nucleophiles) did not
lL of
l

96
J. Ismaili et al. / Steroids 77 (2012) 91–99
Scheme 1. Reagents and conditions: (a) MeSO2Cl, Et3N, toluene, 0 °C, 1.5 h; KHCO3, acetone/H2O 3:1, reflux, 6 h; Jones’ reagent, acetone, 0 °C, 5 min. (b) PyrHCl, LiBr, DMA,
reflux, 4 h. (c) K2CO3, K3Fe(CN)6, OsO4 cat., DHQD, CH3SO2NH2, t-BuOH/H2O 1:1, rt, 6 days. (d) CF3CO3H, CHCl3, rt, 2 h. (e) NMO, OsO4 cat., acetone/H2O 9:1, rt, 12 h. (f) H2SO4,
AcOH, reflux, 2 h; Ac2O, pyridine, DMAP, rt, 3 h. (g) KOH, MeOH, rt, 2 h. (h) Ph3P, p-nitrobenzoic acid, DEAD, THF, rt, 24 h; NaOH, MeOH, rt, 14 h. (i) H2, Pd/C 10%, EtOAc, rt,
14 h.
cleanly lead to the desired alcohol 10. Yields were low and separa-
tion was cumbersome. Compound 10 was then taken to prepare 2
other BRs (see Scheme 1). First, the hydroxy group at position 3
was epimerized under standard Mitsunobu conditions [49,50], elic-
brassinolide (1) against MPP+-induced oxidative stress and apopto-
sis [17]. In the present paper, we report that another natural BR,
homocastasterone (2a), and several analogs (2b, 7, 11, and 10)
have neuroprotective potential at a nanomolar concentration
(10ꢁ9 M).
iting 3a steroid 11 with 49% yield. Second, alcohol 10 was cleanly
hydrogenated (Pd/C, H2, EtOAc) to compound 12 (65% yield).
The BRs and analogs we investigated, when used alone at nano-
molar concentration (10ꢁ9 M), did not cause significant cell death
in neuronal PC12 cells except for analog 22S,23S-homobrassino-
lide (3). BRs, such as 24-epibrassinolide (1) and homocastasterone
(2a), were found, at micromolar concentrations, to inhibit growth
and elicit the apoptosis of several human cancer cell lines without
affecting the growth of normal cells [55]. Yu et al. observed the
micromolar toxicity of 22S,23S-homobrassinolide (3), 22R, 23R-
homobrassinolide (8), 22S,23S-homocastasterone (2b) and 22R,
23R-homocastasterone (2a) on human cancer cell lines with the
(22R,23R) isomer shown to be significantly more toxic than the
(22S,23S) isomer [56]. Our study discerned slight toxicity of the
(22S,23S) isomer of homobrassinolide (3) at nanomolar concentra-
tion in neuronal PC12 but not with the (22R,23R) isomer (Fig. 2B).
Strong as our results are with 9 BRs and analogs, we discuss here
the influence of functionalization and stereochemistry of these phy-
tosterols on neuroprotection. First, we noted that oxygenation of the
B ring greatly influenced BRs neuroprotection against MPP+ toxicity.
Indeed, while ketones 2a and 2b possessed neuroprotective activity,
lactones 3 and 8 did not present any (Fig. 2A). Compound 1, differing
by only 1 carbon at position 24 (ethyl vs methyl) from 3 and 8, dis-
played neuroprotection (Fig. 2A). Therefore, a lactone in the B ring
does not completely obviate neuroprotection against MPP+ toxicity.
Second, stereochemistry at positions 22 and 23 did not influence the
neuroprotective effect of homocastasterone. Indeed, natural homo-
casterone 2a and non-natural analog 2b showed similar activity.
3.2. Cytotoxicity measurements
The ability of BRs and analogs to reverse MPP+-induced cytotox-
icity was investigated by LDH colorimetric assay [35,51]. BRs and
analogs, when used alone, did not cause significant cell death of
neuronal PC12 cells compared to the control condition (Ctrl.,
0.0 0.6%) except for analog 3 (11.7 4.4%, Fig. 2B). Treatment
with MPP+ for 24 h induced 28.8 1.6% cell mortality (Fig. 2A).
When neuronal PC12 cells received compounds 1, 2b, 2a, 7, 11 or
10, cell mortality was significantly reduced (8.5 2.4%; 9.7
2.2%; 13.1 2.3%; 11.9 1.5%; 14.6 2.0 and 11.2 3.6%, respec-
tively) 3 h prior to MPP+. Analog 3, BR 8 and analog 12 did not sig-
nificantly obviate MPP+-induced cell death (24.2 3.6%; 22.2
3.5% and 24.7 3.1%, respectively).
4. Discussion
Since BRs are present in a large number of plants, animals,
including humans, consume BRs in their daily diet. 24-Epibrassin-
olide (1) is found, for example, in V. faba [2], and homocastasterone
(2a) is contained, among others, in Chinese cabbage Brassica cam-
pestris [52,53], rice Oryza sativa [52], rye Secale cereale [54], and
green tea Thea sinensis [52,53]. Our recent study demonstrated
the powerful neuroprotective properties of the natural BR 24-epi-

J. Ismaili et al. / Steroids 77 (2012) 91–99
97
Fig. 2. Cytotoxicity measurement in neuronal PC12 cells by colorimetric assay based on LDH activity. The absorbance value obtained for the untreated control was subtracted
from all other values, as described in the Section 2. The data are expressed as percentages of values of untreated control cells and are means SEM n = 3. ⁄p < 0.05, ⁄⁄p < 0.01,
⁄⁄⁄p < 0.001 vs Ctrl. and ꢃꢃp < 0.01, ꢃꢃꢃp < 0.001 vs MPP+. (A) Effect of BRs or analogs against MPP+-induced cytotoxicity. Neuronal cells were pre-treated with BRs, analogs
(10ꢁ9 M) or vehicle for 3 h and then exposed to MPP+(5 mM) or not for 24 h. (B) Neurotoxicity of BRs or analogs alone in neuronal PC12 cells. Neuronal cells were treated with
BRs or analogs (10ꢁ9 M) or vehicle for 27 h.
Third, the impact of lateral chain oxygenation on neuroprotection
was studied with unsaturated compound 7 in which positions 22
and 23 were not hydroxylated. Neuroprotection similar to hydrox-
ylated BRs 2a and 2b was observed (Fig. 2A). Analog 10, which is
unsaturated in 22 and 23, manifested neuroprotection while its sat-
urated counterpart 12 was inactive (Fig. 2A). Neuroprotection
seems to be dependent on the presence of hydroxyl groups or unsat-
uration at positions 22 and 23. Fourth, we analyzed the effect of
hydroxylation of the A ring in positions 2 and 3. Comparison of neu-
roprotection by analogs 7 and 11 demonstrated that the hydroxyl
group in position 2 was not important for neuroprotection against
MPP+-induced toxicity. Stereochemistry at position 3 did not influ-
ence neuroprotection, as observed by comparing neuroprotection of
analog 11 (3S isomer) with analog 10 (3R isomer).
BR family or metabolites might pass through the BBB [61]. Finally,
BRs have an interesting safety profile. 24-Epibrassinolide (1) inno-
cuity was investigated in mice and rats. Acute toxicity (LD50) was
more than 1000 mg/kg for mice by oral administration, and more
than 2000 mg/kg for rats both orally and dermally [7,62]. The ter-
atogenic potential of homobrassinolide 8 was evaluated in Wistar
rats and it was concluded that 8 is non-teratogenic at doses as high
as 1000 mg/kg [63]. With our results demonstrating the neuropro-
tective effect of BRs on neuronal cell cultures and studies showing
possible passage of BRs through the BBB with their interesting
safety profile, these compounds should thus be regarded as novel
molecules in complementary and/or preventive therapies of neuro-
degenerative diseases and should be assessed in animal models of
such disorders.
BRs should be regarded as interesting molecules for the preven-
tion or treatment of neurodegenerative disease. In this study, sev-
eral natural BRs and analogs were found to protect neuronal cells
against MPP+-induced toxicity. In previous experiments, we associ-
ated neuroprotection by 24-epibrassinolide (1) against MPP+-in-
duced toxicity with antioxidative and anti-apoptotic properties
[17]. Also, recent animal and human studies have shown that phy-
tosterols get through the blood–brain–barrier (BBB) and accumu-
late in the brain [57–59]. Whether this affects neurocognitive
functioning and mental well-being in humans was assessed in
hypercholesterolemic individuals, and it was determined that
long-term (85 weeks) use of plant sterols or stanols (2.5 g/d) has
no impact on neurocognitive functioning or mood [60]. Muthur-
aman et al. reported that homobrassinolide modulated gene
expression in the rat brain, indicating that phytosterol from the
Acknowledgments
Financial support from Fonds Québécois de la recherche sur la
nature et les technologies (FQRNT) is acknowledged. We are thank-
ful to Isabelle Rheault from Université du Québec à Montréal for
performing the HRMS.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.steroids.2011.10.009.

98
J. Ismaili et al. / Steroids 77 (2012) 91–99
the enzymes in the electron transport system in mouse brain. J Neurochem
References
1987;48:1787–93.
[32] Schober A. Classic toxin-induced animal models of Parkinson’s disease: 6-
OHDA and MPTP. Cell Tissue Res 2004;318:215–24.
[1] Park K-H, Yokota T, Sakurai A, Takahashi N. Occurence of castasterone,
brassinolide and methyl 4-chloroindole-3-acetate in immature Vicia faba
seeds. Agric Biol Chem 1987;51:3081–6.
[2] Ikekawa N, Nishiyama F, Fujimoto Y. Identification of 24-epibrassinolide in bee
pollen of the broad bean, Vicia faba L. Chem Pharm Bull 1988;36:405–7.
[3] Gamoh K, Omote K, Okamoto N, Takatsuto S. High-performance liquid
chromatography of brassinosteroids in plants with derivatization using 9-
phenanthreneboronic acid. J Chromatogr 1989;469:424–8.
[4] Bajguz A, Tretyn A. The chemical characteristic and distribution of
brassinosteroids in plants. Phytochemistry 2003;62:1027–46.
[5] McMorris TC, Patil PA, Chavez RG, Baker ME, Clouse SD. Synthesis and
biological activity of 28-homobrassinolide and analogs. Phytochemistry
1994;36:585–9.
[6] Mori K, Sakakibara M, Ichikawa Y, Ueda H, Okada K, Umemura T, et al.
Synthesis of (22S,23S)-homobrassinolide and brassinolide from stigmasterol.
Tetrahedron 1982;38:2099–109.
[7] Khripach V, Zhabinskii V, De Groot A. Twenty years of brassinosteroids:
steroidal plant hormones warrant better crops for the XXI century. Ann Bot
2000;86:441–7.
[8] Brosa C, Nusimovich S, Peracaula R. Synthesis of new brassinosteroids with
potential activity as antiecdysteroids. Steroids 1994;59:463–7.
[9] Khripach VA, Zhabinskii VN, de Groot AE. Brassinosteroids: a new class of plant
hormones. San Diego, CA: Academic Press; 1999.
[10] Clutter HG, Yokota T, Adam G. Brassinosteroids: chemistry, bioactivity and
applications. Washington, DC: American Chemical Society; 1991.
[11] Mazorra LM, Nùnez M, Hechavarria M, Coll F, Sànchez-Blanco MJ. Influence of
brassinosteroids on antioxidant enzymes activity in tomato under different
temperatures. Biol Plant 2002;45:593–6.
[12] Ali B, Hayat S, Fariduddin Q, Ahmad A. 24-Epibrassinolide protects against the
stress generated by salinity and nickel in Brassica juncea. Chemosphere
2008;72:1387–92.
[33] Przedborski S, Tieu K, Perier C, Vila M. MPTP as a mitochondrial neurotoxic
model of Parkinson’s disease. J Bioenerg Biomembr 2004;36:375–9.
[34] Cassarino DS, Fall CP, Swerdlow RH, Smith TS, Halvorsen EM, Miller SW, et al.
Elevated reactive oxygen species and antioxidant enzyme activities in animal
and cellular models of Parkinson’s disease. Biochim Biophys Acta
1997;1362:77–86.
[35] Lahaie-Collins V, Bournival J, Plouffe M, Carange J, Martinoli MG. Sesamin
modulates tyrosine hydroxylase, superoxide dismutase, catalase, inducible NO
synthase and interleukin-6 expression in dopaminergic cells under MPP-
induced oxidative stress. Oxid Med Cell Longev 2008;1:54–62.
[36] Thomas B, Saravanan KS, Mohanakumar KP. In vitro and in vivo evidences that
antioxidant action contributes to the neuroprotective effects of the neuronal
nitric oxide synthase and monoamine oxidase-B inhibitor, 7-nitroindazole.
Neurochem Int 2008;52:990–1001.
[37] Hartley A, Stone JM, Heron C, Cooper JM, Schapira AH. Complex I inhibitors
induce dose-dependent apoptosis in PC12 cells: relevance to Parkinson’s
disease. J Neurochem 1994;63:1987–90.
[38] Shang T, Kotamraju S, Kalivendi SV, Hillard CJ, Kalyanaraman B. 1-Methyl-4-
phenylpyridinium-induced apoptosis in cerebellar granule neurons is
mediated
by
transferrin
receptor
iron-dependent
depletion
of
tetrahydrobiopterin and neuronal nitric-oxide synthase-derived superoxide. J
Biol Chem 2004;279:19099–112.
[39] Ryu EJ, Angelastro JM, Greene LA. Analysis of gene expression changes in a
cellular model of Parkinson disease. Neurobiol Dis 2005;18:54–74.
[40] Sakakibara M, Mori K. Facile synthesis of (22R,23R)-homobrassinolide. Agric
Biol Chem 1982;46:2769–79.
[41] McMorris TC, Chavez RG, Patil PA. Improved synthesis of brassinolide. J Chem
Soc Perkin Trans 1 1996:295–302.
[42] Ramirez JA, Teme Centurion OM, Gros EG, Galagovsky LR. Synthesis and
bioactivity evaluation of brassinosteroid analogs. Steroids 2000;65:329–37.
[43] Korzeniewski C, Callewaert DM. An enzyme-release assay for natural
cytotoxicity. J Immunol Methods 1983;64:313–20.
[44] Martin A, Clynes M. Acid phosphatase: endpoint for in vitro toxicity tests. In
Vitro Cell Dev Biol 1991;27A:183–4.
[13] Özdemir F, Bor M, Demiral T, Türkan I. Effects of 24-epibrassinolide on seed
germination, seedling growth, lipid peroxidation, proline content and
antioxidative system of rice (Oryza sativa L.) under salinity stress. Plant
Growth Regul 2004;42:203–11.
[14] Hayat S, Ali B, Hasan SA, Ahmad A. Brassinosteroid enhanced the level of
antioxidants under cadmium stress in Brassica juncea. Environ Exp Bot
2007;60:33–41.
[15] Arora N, Bhardwaj R, Sharma P, Arora HK. 28-homobrassinolide alleviates
oxidative stress in salt-treated maize (Zea mays L.) plants. Braz J Plant Physiol
2008;20:153–7.
[45] Decker T, Lohmann-Matthes ML.
A quick and simple method for the
quantitation of lactate dehydrogenase release in measurements of cellular
cytotoxicity and tumor necrosis factor (TNF) activity. J Immunol Methods
1988;115:61–9.
[46] Gelinas S, Martinoli MG. Neuroprotective effect of estradiol and
phytoestrogens on MPP+-induced cytotoxicity in neuronal PC12 cells.
Neurosci Res 2002;70:90–6.
J
[16] Muthuraman P, Srikumar K. A comparative effect of homobrassinolide and
giberellic acid on lipid peroxidation and antioxidant defense in alloxan
induced diabetic male albino rats. J Enzyme Inhib Med Chem 2009;24:1122–7.
[17] Carange J, Longpre F, Daoust B, Martinoli MG. 24-Epibrassinolide, a phytosterol
[47] Takatsuto S, Ikekawa N. Remote substituent effect on the regioselectivity in
the Baeyer–Villiger oxidation of 5-alpha-cholestane-6-one derivatives.
Tetrahedron Lett 1983;24:917–20.
[48] Sharpless KB, Amberg W, Bennani YL, Crispino GA, Hartung J, Jeong K-S, et al.
The osmium-catalyzed asymmetric dihydroxylation: a new ligand class and a
process improvement. J Org Chem 1992;57:2768–71.
[49] Mitsunobu O, Yamada M, Mukaiyama T. Preparation of esters of phosphoric
acid by the reaction of trivalent phosphorus compounds with diethyl
azodicarboxylate in the presence of alcohols. Bull Chem Soc Jpn
1967;40:2380–2.
[50] Martin S, Dodge JA. Efficacious modification of the Mitsunobu reaction for
inversions of sterically hidered secondary alcohols. Tetrahedron Lett
1991;32:3017–20.
[51] Bournival J, Quessy P, Martinoli MG. Protective effects of resveratrol and
quercetin against MPP(+)-induced oxidative stress act by modulating markers
of apoptotic death in dopaminergic neurons. Cell Mol Neurobiol
2009;29:1169–80.
[52] Ikekawa N, Takatsuto S, Kitsuwa T, Saito H, Morishita T, Abe H. Analysis of
natural brassinosteroids by gas chromatography and gas chromatography-
mass spectrometry. J Chromatogr 1984;290:289–302.
[53] Abe H, Morishita T, Uchiyama M, Takatsuto S, Ikekawa N, Ikeda M, et al.
Occurrence of three new brassinosteroids: brassinone, (24S)-24-ethylbrassinone
and 28-norbrassinolide, in higher plants. Experientia 1983;39:351–3.
[54] Schmidt J, Spengler B, Yokota T, Nakayama M, Takatsuto S, Voigt B, et al.
Secasterone, the first naturally occurring 2,3-epoxybrassinosteroid from Secale
cereale. Phytochemistry 1995;38:1095–7.
[55] Malikova J, Swaczynova J, Kolar Z, Strnad M. Anticancer and antiproliferative
activity of natural brassinosteroids. Phytochemistry 2008;69:418–26.
[56] Misharin AY, Mehtiev AR, Zhabinskii VN, Khripach VA, Timofeev VP, Tkachev
YV. Toxicity of (22R, 23R)-22,23-dihydroxystigmastane derivatives to cultured
cancer cells. Steroids 2010;75:287–94.
from the brassinosteroid family, protects dopaminergic cells against MPP+
induced oxidative stress and apoptosis. J Toxicol 2011;2011:392859.
-
[18] Mariani E, Polidori MC, Cherubini A, Mecocci P. Oxidative stress in brain aging,
neurodegenerative and vascular diseases: an overview.
2005;827:65–75.
J Chromatogr B
[19] Dauer W, Przedborski S. Parkinson’s disease: mechanisms and models. Neuron
2003;39:889–909.
[20] Olanow CW. The pathogenesis of cell death in Parkinson’s disease – 2007. Mov
Disord 2007;22(Suppl. 17):S335–42.
[21] Zhou C, Huang Y, Przedborski S. Oxidative stress in Parkinson’s disease: a
mechanism of pathogenic and therapeutic significance. Ann NY Acad Sci
2008;1147:93–104.
[22] LeWitt PA. Levodopa therapeutics for Parkinson’s disease: new developments.
Parkinsonism Relat Disord 2009;15(Suppl. 1):S31–4.
[23] Rabey JM, Vered Y, Shabtai H, Graff E, Harsat A, Korczyn AD. Broad bean (Vicia
faba) consumption and Parkinson’s disease. Adv Neurol 1993;60:681–4.
[24] Farina E, Piu P, Strinna L. Extraction of L-DOPA from Vicia faba L. and other
plants of the leguminous genera. Boll Soc Ital Biol Sper 1974;50:508–11.
[25] Raguthu L, Varanese S, Flancbaum L, Tayler E, Di Rocco A. Fava beans and
Parkinson’s disease: useful ‘natural supplement’ or useless risk? Eur J Neurol
2009;16:e171.
[26] Apaydin H, Ertan S, Ozekmekci S. Broad bean (Vicia faba) – a natural source of
L-dopa – prolongs ‘‘on’’ periods in patients with Parkinson’s disease who have
‘‘on–off’’ fluctuations. Mov Disord 2000;15:164–6.
[27] Heikkila RE, Hess A, Duvoisin RC. Dopaminergic neurotoxicity of 1-methyl-4-
phenyl-1,2,5,6-tetrahydropyridine in mice. Science 1984;224:1451–3.
[28] Wong SS, Li RH, Stadlin A. Oxidative stress induced by MPTP and MPP(+):
selective vulnerability of cultured mouse astrocytes. Brain Res
1999;836:237–44.
[29] Greene LA, Tischler AS. Establishment of a noradrenergic clonal line of rat
adrenal pheochromocytoma cells which respond to nerve growth factor. Proc
Natl Acad Sci USA 1976;73:2424–8.
[30] Kadota T, Yamaai T, Saito Y, Akita Y, Kawashima S, Moroi K, et al. Expression of
dopamine transporter at the tips of growing neurites of PC12 cells. J Histochem
Cytochem 1996;44:989–96.
[31] Mizuno Y, Sone N, Saitoh T. Effects of 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine and 1-methyl-4-phenylpyridinium ion on activities of
[57] Jansen PJ, Lütjohann D, Abildayeva K, Vanmierlo T, Plösch T, Plat J, et al. Dietary
plant sterols accumulate in the brain. Biochim Biophys Acta
2006;1761:445–53.
[58] Fricke CB, Schroder M, Poulsen M, von Bergmann K, Wester I, Knudsen I, et al.
Increased plant sterol and stanol levels in brain of Watanabe rabbits fed
rapeseed oil derived plant sterol or stanol esters. Br J Nutr 2007;98:890–9.
[59] Heverin M, Bogdanovic N, Lütjohann D, Bayer T, Pikuleva I, Bretillon L, et al.
Changes in the levels of cerebral and extracerebral sterols in the brain of
patients with Alzheimer’s disease. J Lipid Res 2006;45:186–93.

J. Ismaili et al. / Steroids 77 (2012) 91–99
99
[60] Schiepers OJ, de Groot RH, van Boxtel MP, Jolles J, de Jong A, Lütjohann D, et al.
Consuming functional foods enriched with plant sterol or stanol esters for 85
weeks does not affect neurocognitive functioning or mood in statin-treated
hypercholesterolemic individuals. J Nutr 2009;139:1368–73.
[62] Ikekawa N, Zhao Y-J. Application of 24-epibrassinolide in agriculture. ACS
Symp Ser 1991;474:280–91.
[63] Murkunde YV, Murthy PB. Developmental toxicity of homobrassinolide in
Wistar rats. Int J Toxicol 2010;29:517–22.
[61] Muthuraman P, Srikumar K. Induction of hexokinase I expression in normal
and diabetic rats by a brassinosteroid isoform. Eur J Pharm Sci 2010;41:1–9.